This document is downloaded from DR‑NTU (https://dr.ntu.edu.sg)Nanyang Technological University, Singapore.

Daily Maersk’s impacts on shipper’s supplychain inventories and implications for the linershipping industry

Zhang, Abraham; Lam, Jasmine Siu Lee

2014

Zhang, A., & Lam, J. S. L. (2014). Daily Maersk’s impacts on shipper’s supply chaininventories and implications for the liner shipping industry. Maritime Policy & Management,in press.

https://hdl.handle.net/10356/99084

https://doi.org/10.1080/03088839.2013.869364

© 2014 Taylor & Francis. This is the author created version of a work that has been peerreviewed and accepted for publication by Maritime Policy & Management, Taylor & Francis.It incorporates referee’s comments but changes resulting from the publishing process,such as copyediting, structural formatting, may not be reflected in this document. Thepublished version is available at: [DOI://dx.doi.org/10.1080/03088839.2013.869364].

Downloaded on 12 May 2021 07:50:28 SGT

1

Daily Maersk’s impacts on shipper’s supply chain

inventories and implications for the liner shipping

industry 1

Abraham Zhang*, Jasmine Siu Lee Lam

Division of Infrastructure Systems and Maritime Studies, School of Civil and

Environmental Engineering, Nanyang Technological University, Singapore

Corresponding author e-mail: abrahamzhang@ntu.edu.sg

1 The authors wish to stress that this paper is the result of an independent research project.

The research project has not received any funding support from the industry and the

presented analyses are not influenced by any shipping lines.

2

ABSTRACT

The liner shipping industry has long been characterized by a weekly sailing frequency

and schedule unreliability. This research is motivated by the launch of the revolutionary

“Daily Maersk” service in late 2011, which introduced daily departures and “absolute

reliability” in the Asia–North Europe trade lane. This paper analyzes Daily Maersk’s

impacts on a shipper's supply chain inventories and profound implications for the liner

shipping industry as a game changer. The quantitative analyses show that the impact of

more frequent sailings is most significant on a shipper’s cycle stock, while improving

schedule reliability substantially reduces safety stock and pipeline stock. Daily Maersk is

most valuable for products that have high value density, high inventory holding cost ratio,

low demand variability and high service level requirement. These findings imply that the

trend of liner alliance/merger/acquisition is likely to continue or even accelerate as

shipping lines consolidate fleet capacity to offer more frequent sailings. Rival carriers

may step up their involvement in terminal operations to improve schedule reliability.

They also need to rethink about their service level targets and clearly define their

preferred customer segments.

Keywords: Daily Maersk; liner shipping; sailing frequency; schedule reliability; supply

chain management; maritime supply chain

3

1. Introduction

The liner shipping industry has long been characterized by weekly departures and

schedule unreliability (Vernimmen, Dullaert, and Engelen 2007; Notteboom and

Rodrigue 2008). To ship containers more frequently, a shipper often has to engage

several ocean carriers that have different departure dates in a week. In this case, a shipper

cannot bargain as much discount with any single carrier due to scattered cargo volume.

Its logistics management also becomes more complicated, not just because it has to deal

with multiple carriers, but also because of their schedule unreliability. According to the

statistics provided by Drewry (2010), schedule delay in the container liner shipping

industry averaged 32% to 54% from December 2005 to June 2010. Such low schedule

reliability makes it very difficult for shippers to practice efficient just-in-time

production/distribution (Lam and Van De Voorde 2011). As a result, shippers have to

keep more safety stock, or from time to time use more expensive transportation modes to

expedite cargoes (Vernimmen, Dullaert, and Engelen 2007).

This study is motivated by the launch of “Daily Maersk” in late 2011 by Maersk Line,

the world's largest container shipping company. Daily Maersk serves the Asia–North

Europe trade lane. It has daily cut-offs at four Asian ports and guarantees cargo

availability after a certain number of days at any one of the three North European ports,

as indicated in Figure 1. The service claims “absolute reliability” and promises to pay a

penalty for late arrivals (Maersk Line 2011). It sharply contrasts a standard liner shipping

service that has a weekly sailing frequency and low schedule reliability.

Source: adapted from Maersk Line (2011)

Figure 1. Daily Maersk port calls and transportation times

Daily Maersk has created a stir in the liner shipping industry. Most shippers have

applauded the revolutionary service offering as it helps cut supply chain inventories,

which constitute a major cost component in maritime logistics (Leach 2011). Some have

even praised Daily Maersk as the most positive initiative coming out of the shipping

industry in the past decade (Maersk Line 2011). To counter Daily Maersk, rival carriers

have quickly restructured their alliances to offer more frequent sailings on Asia–Europe

trade lanes (Barnard 2011; Szakonyi 2011). The shipping market expects more profound

changes in the years to come because of Daily Maersk.

4

This paper analyzes Daily Maersk’s impacts on a shipper's supply chain inventories

and implications for the liner shipping industry as a game changer. To the best of our

knowledge, this is the first scholarly research about Daily Maersk. It is well-established

in the literature that maritime transport services should be studied in the context of the

supply chain that they are embedded in, as supply chain management is now widely

practiced by industry professionals (Robinson 2002; Christopher 2005; Carbone and

Gouvernal 2007; Cahoon and Notteboom 2008; Zhang and Huang 2012). Many research

papers have studied ports from a supply chain perspective, for example, the value

propositions and strategic roles of ports/terminals (Robinson 2002; Paixao and Marlow

2003; Robinson 2006; Mangan, Lalwani, and Fynes 2008; Rodrigue and Notteboom 2009;

Zhang, Lam, and Huang 2013), port/terminal integration in the supply chain (Carbone

and Martino 2003; Panayides and Song 2008; Song and Panayides 2008; Panayides and

Song 2009; Tongzon, Chang, and Lee 2009), port performance measurements (Marlow

and Paixao 2003; Bichou and Gray 2004), and port competition in supply chain systems

(Lam and Yap 2011a, 2011b). Surprisingly, very few papers (Vernimmen, Dullaert, and

Engelen 2007; Notteboom and Rodrigue 2008; Saldanha et al. 2009; Lam and Van De

Voorde 2011) have adopted a supply chain perspective to study liner shipping services.

This paper aims to narrow the research gap by achieving the following three objectives.

First, it performs a case study of Daily Maersk to quantify the impacts of improving

sailing frequency and schedule reliability on a shipper’s supply chain inventories. Second,

it identifies the product segment that Daily Maersk is most valuable to. Third, it discusses

the implications of Daily Maersk for the liner shipping industry based on the analytical

results and findings. This study offers insights for ocean carriers who strive to deliver

superior customer values in a highly competitive business environment.

The rest of this paper is organized as follows. Section 2 reviews relevant literature and

defines the analytical methodology in the context of a representative maritime supply

chain. Section 3 explains problem settings and data. Section 4 presents case study results

and analyses. Section 5 discusses Daily Maersk’s implications for the liner shipping

industry. Section 6 concludes the research.

2. Methodology

Some transport related studies examined the impact of service attributes on supply chain

inventories. Blauwens et al. (2003), Dullaert et al. (2007) and Blauwens et al. (2006)

developed analytical models to compare the total logistics costs of using different

transport modes and service frequencies at inland transport stage. Bertazzi and Speranza

(1997; 1999) built mixed integer programming models to minimize total logistics costs

with given transport frequencies. Constable and Whybark (1978) developed a

mathematical model and exact and heuristic procedures for jointly determining inventory

control policies and transport decisions. Allen et al. (1985) used an analytical model to

illustrate the effects of transit time reliability on optimal inventory control policies.

Tyworth and Zeng (1998) developed an analytical model to evaluate the effects of transit

time performance on logistics cost and service. All these analytical studies showed that

improving transport service frequency or reliability allows shippers to reduce supply

chain inventories. Several empirical studies also affirmed that service frequency

(Zamparini, Layaa, and Dullaert 2011) and reliability (Train and Wilson 2008) are

5

importance freight transport service attributes. However, their analyses were primarily

based on inland transport modes and did not consider the distinctive attributes of liner

shipping.

Notteboom (2006) was among the first to highlight the importance of schedule

reliability in liner shipping. Ocean carriers’ schedule unreliability was identified as a key

issue impeding maritime supply chain efficiency (Vernimmen, Dullaert, and Engelen

2007; Notteboom and Rodrigue 2008). Through a simulation study, Saldanha, et al. (2009)

suggested that more reliable liner shipping services can render shippers significant

inventory cost savings. However, little research has been conducted on the combined

effect of improving liner shipping sailing frequency and schedule reliability, which

distinguishes Daily Maersk from a standard service.

Adapting the works of (Allen, Mahmoud, and McNeil 1985; Tyworth and Zeng 1998;

Blauwens et al. 2003; Blauwens et al. 2006; Dullaert et al. 2007; Saldanha et al. 2009),

we develop an analytical model to calculate inventories required in a shipper’s maritime

supply chain. The analytical model provides a theoretical basis to quantify the inventory

cost savings that Daily Maersk can deliver for its customers . It considers three types of

inventories: pipeline stock, cycle stock and safety stock. Pipeline stock refers to

inventories that are in-transit at varies transport stages. Cycle stock at the receiving

location is the portion of inventory available for the average demand during cargo arrival

intervals. Safety stock at the receiving location is used to cover demand variations

(Liberatore and Miller 1995; Coyle, Bardi, and Langley 2009). The following notations

are defined.

V = value of a container load in a twenty-foot equivalent unit (TEU) (USD)

D = demand per day, a random variable with mean D and standard deviation D

(TEUs per day)

CovD = coefficient of variation of demand ( DD / )

R = annual demand (TEUs)

Q = shipment size (TEUs)

ih = holding cost ratio for pipeline stock (percent per year)

wh = holding cost ratio for cycle and safety stocks (percent per year)

L = door to door transport lead time, a random variable with mean L and standard

deviation L (days)

K = safety stock factor corresponding to a pre-specified service level target

s = number of sailings per week

This study considers a hypothetical but representative case of liner shipping from

Shenzhen Yantian port in China to the port of Rotterdam in Netherlands. The case

involves a manufacturer seller, a buyer (shipper) and an ocean carrier. The seller exports

containerized cargoes based on Incoterms Ex Works (EXW) from Dongguan, Guangdong

province, China. Note that several other Incoterms are also widely used in maritime

transport, for example, Free on Board (FOB), and Cost, Insurance and Freight (CIF)

(Cahoon and Notteboom 2008). This case study chooses EXW trade term as it gives a

shipper most control over a maritime supply chain, and therefore allows the quantitative

analyses to unveil the impacts of Daily Maersk more completely. The buyer has a

6

regional warehouse about 2 hours trucking distance from the port of Rotterdam. The

buyer has a contract with the ocean carrier to ship all cargoes from the seller to the

regional warehouse. On average, the buyer has more than one container to transport each

day and all products are shipped in full containers. Order processing cost is not

considered in the analyses as it is often a very small portion of total logistics cost

(Saldanha et al. 2009).

Average pipeline stock (PS) is calculated as (Leachman 2008)

LDLR

PS

365

(1)

On average, half of the shipment size is in cycle stock (CS) (Blauwens et al. 2003):

2

QCS (2)

A year is approximately 52 weeks. The buyer ships containers regularly according to

the sailing frequency of the ocean carrier to minimize cycle stock. Equation (2) can be

rewritten as follows:

ss

RCS D

2

7

522

(3)

Broadly speaking, there are three approaches for setting safety stocks, namely the

“time supply” approach, the “shortage costing” approach and the “service level”

approach (Silver, Pyke, and Peterson 1998). The first approach sets safety stock equal to

a certain time of supply. In reality, some shippers keep more than one week safety stock

because of the schedule unreliability of liner shipping services. However, if we follow

this approach, the results may exaggerate the benefits of Daily Maersk. The second

approach minimizes the total of shortage cost and inventory holding cost. This approach

is difficult to implement in practice because it is usually hard to quantify shortage costs

(Vernimmen, Dullaert, and Engelen 2007). The third approach bypasses the problem. It

minimizes inventory holding cost subject to a service level target.

The analytical methodology adopted in this paper follows the “service level” approach

to calculate safety stock. It measures service level as the probability of no stock-out per

replenishment cycle. An alternative service measure is fill rate, i.e. a fraction of demand

to be satisfied from stock on hand. These two service measures lead to different

formulations for safety stock calculation. This study employs the first service measure as

it is most often used in practice (Dullaert et al. 2007).

To compute safety stock, it is essential to first derive the standard deviation of demand

during lead time (DDLT). We assume that lead time is independent of demand and

demand itself is not auto-correlated (Vidal and Goetschalckx 2000). In this case, the

standard deviation of DDLT can be computed as follows (Silver, Pyke, and Peterson

1998; Vernimmen, Dullaert, and Engelen 2007):

222222222

LLDLDDLLDDLDDLT CovDCovD (4)

7

After obtaining the standard deviation of DDLT, the level of safety stock (SS)

corresponding to a pre-specified service level target can be calculated as follows:

22

LLDDDLT CovDKKSS (5)

Assuming demand follows a normal distribution, the value of K can be easily obtained

from any statistical handbook. For example, the value of K corresponding to a stock-out

risk of 5% (i.e. 95% service level) is equal to 1.64. It means that a safety stock of 1.64

times the standard deviation of DDLT is required to ensure a 95% service level. For

service levels of 97% and 99%, their corresponding K factors amount to 1.88 and 2.33

respectively (Blauwens et al. 2003). Note that the values of K may be different for

demand patterns not following a normal distribution, for example, Erlang and Poisson

distributions. However, equation (5) is still applicable for the calculation of safety stock.

Based on equations (1), (3) and (5), the quantity of total supply chain inventories (SCIQ)

can be obtained by summarizing pipeline stock, cycle stock and safety stock.

22

2

7LLD

DLD CovDK

sSCIQ

(6)

The cost of holding these supply chain inventories (SCIC) can be calculated as follows:

VhCovDKs

VhSCIC wLLDD

iLD )2

7(

22

(7)

From equations (6) and (7), it can be observed that supply chain inventories are linear

to D . This implies that demand volume can be filtered if we measure inventories by

days of demand.

3. Problem settings and data

Figure 2 illustrates door-to-door transport lead time segments for a marine container

shipped from Asia to North Europe. The total transport lead time is partitioned into seven

segments. The seven segments can be grouped into three transport stages: at export

country in Asia (stage 1), in ocean transit (stage 2), and at import country in North

Europe (stage 3). A standard liner shipping service gives scheduled ocean transit time,

while Daily Maersk guarantees transportation time. Maersk Line (2011) defines

transportation time as the time taken from the container terminal cut-off at origin until the

cargo becomes available for pick-up at destination. In contrast, ocean transit time only

includes the time from when vessel departs at origin until it arrives at destination.

8

Source: adapted from Saldanha, et al. (2009)

Figure 2. A maritime container’s door-to-door transport lead time segments

This research defines the following liner shipping services with different sailing

frequencies and schedule reliability. Sailing frequency is based on the number of

departures from a same port. Schedule reliability is defined as the percentage of times

that a ship arrives at a port on the scheduled day or on the day immediately before the

scheduled day of arrival (Drewry 2010). Service I represents a standard service that has a

weekly sailing frequency. Its schedule reliability is 70%, the industry average in late

2011 when Daily Maersk was launched (Drewry 2012). Service II only improves sailing

frequency to daily. Service III only improves schedule reliability to 99%, which is the

service level that Daily Maersk achieved in the first two months of its official operation

(Maersk Line 2011). Daily Maersk improves not only sailing frequency but also schedule

reliability.

Service I: weekly sailings, 70% schedule reliability

Service II: daily sailings, 70% schedule reliability

Service III: weekly sailings, 99% schedule reliability

Daily Maersk: daily sailings, 99% schedule reliability

According to a detailed survey of Drewry (2010), Figure 3 shows the pattern of ocean

transit time in liner shipping. A histogram is used to approximate the transit time

distribution, with the horizontal axis representing days of deviation from schedule and the

vertical axis representing occurrence frequency. Following the format used in (Saldanha

et al. 2009), the histogram can be described numerically as “Histogram Min = -3, Max =

9, Freq = {1, 2, 8, 45, 19, 9, 5, 4, 2, 1, 2, 1, 1}”. In numerical expression, the values of

“Min” and “Max” set the lower and upper bounds of transport lead time measured in days.

The values of “Freq”, a set of numbers in the bracket, represent the occurrence

frequencies of various transport lead times that equally divide the transport lead time

range.

3A

Port-dwell

till cargo

ready for

pick-up

3B

Port

waiting,

customs

clearance

1A

Export

drayage

1C

Port-dwell

from

terminal

cut-off time

2

Ocean

transit

1B

Port

waiting,

customs

clearance

3C

Import

drayage

Daily Maersk’s transportation time

Export Country in Asia Import Country in North Europe

9

Source: Drawn by authors based on data from Drewry (2010)

Figure 3. Number of days deviation from scheduled day of arrival

For liner shipping services that have different schedule reliability, we assume their

days of early/delayed arrival follow a similar pattern as depicted in Figure 3. Ocean

transit time distribution for Services I and II, whose schedule reliability is 70%, can be

approximated as “Histogram Min = 24, Max = 34, Freq = {2, 10, 60, 12, 6, 3, 3, 1, 1, 1,

1}”. Based on the histogram approximation, it is easy to calculate their ocean transit time

as 26.57 days on average with a variance of 2.73 days2. Similarly, Service III's ocean

transit time distribution is approximated as “Histogram Min = 25, Max = 27, Freq = {15,

84, 1}”. The average ocean transit time is 25.86 days with a variance of 0.14 days2. Daily

Maersk's transportation time distribution is approximated as “Histogram Min = 29, Max

= 31, Freq = {15, 84, 1}”. Its average value is 29.86 days with a variance of 0.14 days2.

Data on inland transport lead times were estimated in consultation with an established

freight forwarder that has customers shipping via Shenzhen Yantian port regularly. The

freight forwarder is believed to be unbiased because it is neutral and is not affiliated with

any shipping lines including Maersk Line. Although secondary data are not available for

the validation of detailed segment lead times, total inland transport lead times at import

and export ends are found to be in line with those reported in the recent literature

(Saldanha, Russell, and Tyworth 2006; Leachman 2008). The appendix gives detailed

data for histogram approximation.

Table 1 summarizes all transport lead time means (the first numbers in brackets) and

variances (the second numbers in brackets). According to the illustration in Figure 2, the

maritime supply chain includes three transport stages. Assuming their lead times are

independent from each other, the mean and variance of total transport lead time are the

sum of those of the three transport stages.

Table 1. Data of transport lead time means and variances

Transport stage 1 Transport stage 2 Transport stage 3 Total transport

lead time

10

Service I (4.70, 1.04) (26.57, 2.73) (2.50, 0.55) (33.77, 4.32)

Service II (3.25, 0.33) (26.57, 2.73) (2.50, 0.55) (32.32, 3.61)

Service III (4.43, 0.51) (25.86, 0.14) (2.50, 0.55) (32.79, 1.20)

Daily Maersk (0.44, 0.08) (29.86, 0.14) (1.53, 0.49) (31.82, 0.71)

Note: For Daily Maersk, transport stage 1 includes only segments 1A and 1B as

illustrated in Figure 2. Its transport stage 2 includes segments 1C, 2 and 3A, and transport

stage 3 includes segments 3B and 3C.

Port-of-departure dwell time is influenced by factors including carriers’ cut-off time

requirements, sailing frequency, vessel schedule reliability and customs clearance.

Carriers often require shippers to deliver containers to a port-of-departure at least two

days before the booked vessel’s scheduled arrival. Table 1 shows that improving sailing

frequency reduces both the mean and variance of port-of-departure dwell time. This is

because shippers tend to keep longer buffer time when sailings are infrequent as missing

a vessel may result in long waiting time for the next available vessel. To further reduce

the variance of port-of-departure dwell time, however, schedule reliability must be

improved. This is because schedule unreliability is a cause of port-of-departure dwell

time variability. Besides vessel arrival delays, uncertainties associated with customs

clearance play a key part to prolong port-of-departure dwell time. In mainland China,

export containers must arrive one day in advance for customs clearance and inspection. In

the case of open-box inspection, actual time taken might be up to two days or even more.

For this reason, freight forwarders often require cargo owners to deliver containers two to

three days earlier than ocean carriers’ terminal cut-off times. This causes port-of-

departure dwell time to typically take four to five days in China if there is no delay of

vessel arrival. At the port-of-entry, it is assumed that port dwell time is not affected by

liner’s schedule unreliability.

In the base case scenario, the coefficient of variance of demand (CovD) is set as 0.075.

Such a demand variance means that approximately 95% of random daily demands would

fall within ±15 % of the mean for a bell-shaped distribution (Saldanha et al. 2009).

Holding cost ratio for cycle and safety stocks ( wh ) is conservatively set as 20%per year

(Huang, Zhang, and Liang 2005; Leachman 2008). In all scenarios, holding cost ratio ih

for pipeline stock is set as 10% lower than wh because pipeline stock does not incur

warehousing cost (Saldanha et al. 2009).

4. Results and analyses

4.1. Impacts of improving sailing frequency and schedule reliability

This section employs the methodology defined in Section 2 to quantify the impacts of

improving liner shipping sailing frequency and schedule reliability. As mentioned

previously, a superior transport service allows a shipper to reduce supply chain

inventories. Table 2 shows required supply chain inventories calculated based on

equations (1), (3), (5) and (6). The definitions of liner shipping services follow those

given in Section 3. The shipper is assumed to keep a 97% service level (SL).

11

Table 2. Supply chain inventories (measured by days of demand)

Service

I

Service

II

Service

III

Daily

Maersk

Cycle stock 3.50 0.50 3.50 0.50

Pipeline stock 33.77 32.32 32.79 31.82

Safety stock (SL=97%) 3.99 3.66 2.21 1.77

Total Inventory (SL=97%) 41.26 36.48 38.5 34.09

It is apparent that cycle stocks are the lowest for Service II and Daily Maersk because

of their daily sailings. In comparison with a standard weekly service, daily sailings allow

a shipper to transport cargoes in much smaller batches. Consequently, cycle stock can be

reduced from 3.5 to 0.5 days.

Pipeline stock measured by days of demand is equal to the mean of total transport lead

time. Due to long transit time in maritime logistics, pipeline stock is the largest

component of supply chain inventories. For the service types included in Table 2,

considerable differences can be observed in the required pipeline stocks. In comparison

with a standard service, Service II reduces pipeline stock from 33.77 to 32.32 days. This

is because shippers no longer need to keep as much buffer time at port-of-departure when

the consequence of missing a booked vessel is less severe with more frequent sailings.

Improving schedule reliability, Service III can reduce pipeline stock to 32.79 days

because of less transport delays. Daily Maersk leads to the lowest pipeline stock of 31.82

days because of the combined effect of improved sailing frequency and schedule

reliability. Although it is not within the scope of this study, it should be acknowledged

that shippers’ pipeline stocks have been adversely affected by carriers’ implementation of

slow steaming in recent years.

Safety stock is only marginally reduced by improving sailing frequency alone from

weekly (Service I) to daily (Service II). This is because more frequent sailings lead to

lower variance of port-of-departure dwell time. However, the variance of port-of-

departure dwell time cannot be much reduced due to ocean vessels’ low schedule

reliability. As mentioned previously, schedule unreliability is a major cause of prolonged

port-of-departure dwell time. Improving schedule reliability alone from 70% (Service I)

to 99% (Service III) can reduce safety stock from 3.99 to 2.21 days. Daily Maersk further

reduces safety stock requirement to 1.77 days by improving both sailing frequency and

schedule reliability.

In summary, the impact of more frequent sailings is most significant on cycle stock,

followed by pipeline stock and safety stock. Improving schedule reliability reduces safety

stock and pipeline stock, but has no impact on cycle stock. Daily Maersk, which

improves both sailing frequency and schedule reliability, can reduce all three supply

chain stocks significantly, and thus achieve maximal inventory reductions.

Table 3. Supply chain inventory cost savings over a standard service

Service II Service III Daily Maersk

SL=95% 16.8% 8.6% 24.8%

12

SL=97% 16.6% 9.3% 25.4%

SL=99% 16.3% 10.7% 26.6%

Table 3 compares supply chain inventory cost savings resulting from improving

sailing frequency or/and schedule reliability. Inventory costs are calculated based on

equations (7). In comparison with a standard service (Service I), daily sailings (Service

II) can help cut inventory costs by 16.6% when the shipper keeps a 97% service level. If

the sailing frequency is unchanged as weekly, improving schedule reliability from 70%

(Service I) to 99% (Service III) can save inventory costs by 9.3%. Daily Maersk can

enable much greater inventory cost savings of 25.4%.

A shipper’s service level target influences potential inventory cost savings. Table 3

shows that when a shipper sets a higher service level target, Services II delivers slightly

less percent savings. In contrast, Services III and Daily Maersk can achieve greater

percent savings. This is because a larger portion of safety stock is required at a higher

service level to cover demand and transport lead time variations. Service II is

disadvantaged at a higher service level as it does not directly tackle schedule

unreliability, as in the case of Service III and Daily Maersk.

It is worthwhile to take note that the involvement of multiple stakeholders and

authorities causes the maritime supply chain to become very fragmented, which is

reflected in the many lead time segments as illustrated in Figure 1. Consequently,

shippers suffer from high inventory costs. To fully integrate the maritime supply chain,

not only shipping lines, terminal operators, shippers, freight forwarders and inland

transport service providers must work together, but also port authorities and customs.

Shipping lines have been extending upstream and downstream chain control (Notteboom

and Rodrigue 2008). However, a portion of inventory costs in the maritime supply chain

will remain fixed due to customs and government regulations unless they are simplified

and relevant procedures are streamlined.

4.2. Shipping rate premiums

Table 4. Daily Maersk's shipping rate premiums

Product value/TEU

Holding cost ratio wh

$10,000 $50,000 $90,000 $130,000 $170,000

20% $34 $170 $306 $441 $577

40% $73 $366 $659 $952 $1,245

60% $113 $563 $1,013 $1,463 $1,914

Table 4 presents Daily Maersk's shipping rate premiums in relation to product value

density and holding cost ratio wh . Shipping rate premium is defined as equal to the

inventory cost savings resulting from the use of a superior service in comparison with a

standard service (Saldanha et al. 2009). Rate premiums serve as a gauge how much more

an ocean carrier could charge, or a shipper may be willing to pay, for a superior liner

13

service. It is measured by USD per TEU. For a fully loaded forty-foot container, the

amount of rate premium should double. The service level requirement of shippers is

assumed to be 97%.

It is apparent that Daily Maersk's shipping rate premiums vary substantially by

product value density and holding cost ratio. Note that these two parameters are largely

determined by product nature (Lovell, Saw, and Stimson 2005). Industrial raw materials

and commodity-type products often have a low holding cost ratio of about 20%.

However, the ratio may be much higher for perishable, fashionable and electronic

products that have a short product life cycle (Leachman 2008). . For a product that has a

low value of $10,000 per TEU, Daily Maersk's rate premiums are insignificant. However,

for products whose values are $90,000 and $170,000 per TEU, the rate premiums amount

to $306 and $577 respectively when the holding cost ratio is 20%. In the case of a 60%

holding cost ratio, which may be applicable for dairy products and consumer electronics,

Daily Maersk's rate premiums are equal to $1,013 and $1,914 respectively.

The inventory cost savings of improving sailing frequency and schedule reliability, as

quantified above for the case of Daily Maersk, have been supported by evidences in the

industry. According to Page (2010), many shippers have now made their priorities clear

to ocean carriers: service, reliability and price, in that order. Near to the one year

anniversary of Daily Maersk, Maersk Line (2012) conducted a survey of 175 customers.

Most of them stated that they had saved either time or money. 42% responded that they

had saved logistics costs as a direct result of using Daily Maersk. The benefits are

especially substantial and tangible for products that have high value density and high

inventory holding cost ratio. A specific product example is laptop computers, which are

of high values and whose values quickly depreciate over time as new and improved

products continuously hit the market (Rao 2012).

4.3. Sensitivity analysis of demand variance

14

Figure 4. Supply chain inventory cost savings over a standard service

Figure 4 presents the results of sensitivity analysis of demand variance. It is less

beneficial to use Daily Maersk when demand variance is greater. This is because more

safety stock is required to ensure the same service level when demand variance increases.

This undermines the comparative advantage of Daily Maersk as safety stock reduction is

one of its key benefits. Service II is not sensitive to changes in demand variance, because

safety stock reduction only makes marginal contributions to its inventory cost savings.

Nevertheless, wide gaps still exist between Daily Maersk’s and Service II’s inventory

cost savings. This finding is in line with that of Saldanha, et al. (2009). They suggested

that cost savings resulted from transit time reliability would not disappear until CovD

approaches 2.0, which is very rare. Consistent with the findings presented in Section 4.1,

Daily Maersk is more beneficial for products that have a higher service level requirement.

5. Implications for the liner shipping industry

Based on the results and analyses presented above, we can advance two general

propositions. The first general proposition is that increasing sailing frequency can help

shippers reduce cycle stock most significantly, followed by pipeline stock and safety

stock. Improving schedule reliability helps substantially reduce safety stock and pipeline

stock, but has not impact on cycle stock. Maximal inventory cost savings can only be

achieved by improving both sailing frequency and schedule reliability, as in the case of

Daily Maersk. The second general proposition is that Daily Maersk is most valuable for

products that have high value density, high inventory holding cost ratio, low demand

variability and high service level requirement. There are several implications that can be

derived from these general propositions for the liner shipping industry.

First, it is indeed necessary for rival carriers to improve sailing frequency to compete

against Daily Maersk. More frequent sailings offered by rival carriers are supported by

two important changes in their alliance structure to consolidate fleet capacities. One

change was the birth of the G6 Alliance from merging the New World Alliance

(NOL/APL, MOL, HMM) and the Grand Alliance (Hapag-Lloyd, NYK, OOCL).

Another change was the establishment of the MSC/CMA CGM Alliance. As a result of

these two changes, 90% market share in the Far East/Europe route is now controlled by

the big four powers (Moon 2012), namely Maersk Line, the G6 Alliance, the MSC/CMA

CGM Alliance and the CKYH Alliance (COSCO, K-Line, Yang Ming, Han Jin). This

shows that Daily Maersk is indeed a market-changing service. With shippers’ expectation

now being raised to more frequent sailings, the trend of liner alliance/merger/acquisition

(Midoro and Pitto 2000; Slack, Comtois, and McCalla 2002; Lam, Yap, and Cullinane

2007) is likely to continue or even accelerate.

Second, it is crucial for carriers to improve schedule reliability to enable shippers

further reduce supply chain inventories. However, it would be more challenging to

improve schedule reliability than to increase sailing frequency. Schedule unreliability has

been mainly caused by factors uncontrollable to carriers, for example, port/terminal

congestion, port/terminal productivity below expectations (Notteboom 2006). To secure

capacity in key ports in their service schedules, carriers have become increasingly

involved in terminal operations. Just to name a few, Maersk Line, MSC, CMA CGM,

15

COSCO and Evergreen have all been very active in making strategic investment in

container terminals (Slack and Fremont 2005; Notteboom 2006; Vernimmen, Dullaert,

and Engelen 2007; Pawlik et al. 2011). In the case of Maersk Line, its parent company

AP Moller-Maersk operates a large number of container terminals worldwide through its

subsidiary APM Terminals (Vernimmen, Dullaert, and Engelen 2007). This enables

Maersk Line to use a global network of dedicated hub terminals (Fremont 2007) to

achieve the highest schedule integrity in the liner shipping industry (Notteboom and

Rodrigue 2008). As Daily Maersk attracts shippers by a very high service level, rival

carriers may step up their involvement in terminal operations to improve schedule

reliability.

Third, the rate premium of Daily Maersk is much more significant for products that

have high value density, high inventory holding cost ratio, low demand variation and

high service level requirement. This implies that market segmentation has become

important in the liner shipping industry. In the past, most shippers regarded all liner

services as identical (Saldanha et al. 2009) and the industrial competition focused on cost.

To a large extent, the liner shipping industry ignored customer value creation in other

areas, for example, transit time and service quality (Robinson 2006; Lee and Song 2010).

As Daily Maersk breaks the industrial norm, shippers have quickly recognized the value

of more frequent sailings and schedule reliability. Rival carriers now have to rethink

about their service level targets and the strategic positioning of their service offerings.

In summary, Daily Maersk has made profound impact on the liner shipping industry.

In the Asia–Europe trade lane, industrial competition has been much intensified since the

announcement of Daily Maersk. Although Maersk Line did not give a time line, it stated

that similar services should be rolled out worldwide. When that happens, the global liner

shipping industry may be profoundly reshaped. Coupled with overcapacity issues caused

by sluggish demand, the liner shipping industry may face a big reshuffle in the years to

come.

6. Conclusions

This paper employs a supply chain perspective to analyze Daily Maersk’s impacts on a

shipper's inventory management and its implications for the liner shipping industry. In

the past few decades, a standard liner shipping service has had a weekly sailing frequency

and low schedule reliability. In late 2011, Maersk Line launched its market-changing

Daily Maersk service that offers a daily sailing frequency and promises “absolute

reliability” in the Asia–North Europe trade lane. Daily Maersk has quickly gained

popularity among shippers as it helps significantly reduce supply chain inventories,

which constitute a major cost component in maritime logistics. To prevent loss of market

share, rival carriers have quickly changed their alliance structure to provide more

frequent sailings. In light of the far-reaching impact of Daily Maersk, this paper

quantifies Daily Maersk’s impacts on a shipper's supply chain inventories and discusses

its profound implications for the liner shipping industry.

This paper makes several important contributions. First, it is believed to be the first

research study about Daily Maersk. Daily Maersk is revolutionary and has been regarded

16

as among the most positive initiative coming out of the liner shipping industry in the past

decade. It warrants further study. Second, it employs a supply chain perspective to

quantify Daily Maersk’s benefits, which are evidential in a shipper’s inventory cost

savings. So far, limited research has adopted a supply chain perspective to study liner

shipping services. Third, the analytical results show that increasing sailing frequency

helps reduce cycle stock most significantly, followed by pipeline stock and safety stock,

while improving schedule reliability helps reduce safety stock and pipeline stock. For the

representative case given in the study, increasing sailing frequency from weekly to daily

can help cut supply chain inventory costs by 16.6%. If schedule reliability is also

improved to 99% as in the case of Daily Maersk, inventory cost savings could amount to

25.4%. Fourth, the analyses identify that Daily Maersk is most valuable for products that

have high value density, high inventory holding cost ratio, low demand variance and high

service level requirement. Last but not least, it discusses the implications of Daily Maersk

for the liner shipping industry. The trend of liner alliance/merger/acquisition is likely to

continue or even accelerate. Rival carriers may need to become more involved in terminal

operations to improve their schedule reliability. They also have to rethink about their

service level targets and the strategic positioning of their service offerings.

This research has its limitations. The quantitative analyses conducted in this paper are

theoretical. Although the quantified benefits are believed to be conservative and realistic,

shippers need to implement appropriate inventory control policies to realize the potential

inventory cost savings. It would be beneficial to conduct a comprehensive empirical

study to validate actual inventory cost savings of shippers who have switched to Daily

Maersk. In addition, we are extending the research to analyze the implications of Daily

Maersk for the port industry.

Acknowledgements

The authors would like to thank the industry professionals who provided help on the

data. They appreciate constructive comments offered by the editor Prof. Kelvin Li,

associate editor Prof. Wesley Wilson and two anonymous reviewers. Their comments

have greatly helped improve the paper. This research was partially funded by Singapore

MOE AcRF NTU Ref: RF20/10.

References

Allen, W. Bruce, Mohamed M. Mahmoud, and Douglas Mcneil. 1985. The Importance of

Time in Transit and Reliability of Transit Time for Shippers, Receivers, and Carriers.

Transportation Research Part B 19B (5): 447-456.

Barnard, Bruce. 2011. "Maersk Adds Three European Ports to Daily Maersk." The

Journal of Commerce. Accessed 15 February 2012. http://www.joc.com/container-

lines/maersk-adds-three-european-ports-daily-maersk.

Bertazzi, Luca, and Maria Grazia Speranza. 1999. Inventory Control on Sequences of

Links with Given Transportation Frequencies. International Journal of Production

Economics 59 (1-3): 261-270.

Bertazzi, Luca, and Maria Grazia Speranza. 1997. Minimization of Logistic Costs with

Given Frequencies. Transportation Research Part B 31 (4): 327-340.

17

Bichou, Khalid, and Richard Gray. 2004. A Logistics and Supply Chain Management

Approach to Port Performance Measurement. Maritime Policy & Management 31 (1):

47-67.

Blauwens, G., S. Janssens, B. Vernimmen, and Frank Witlox. 2003. "A Logistical

Analysis of Transport Frequency." In Across the Border: Building Upon a Quarter

Century of Transport Research in the Benelux, edited by Dullaert, W., Bart A.M.

Jourquin and Jacob B. Polak, 69-88. Antwerp, Belgium: De Boeck

Blauwens, Gust, Nico Vandaele, Eddy Van De Voorde, Bert Vernimmen, and Frank

Witlox. 2006. Towards a Modal Shift in Freight Transport? A Business Logistics

Analysis of Some Policy Measures. Transport Reviews 26 (2): 239-251.

Cahoon, Stephen, and Theo E. Notteboom. 2008. "Port Marketing Tools in a Logistics

Restructured Market Environment: The Quest for Port Royalty." Paper presented at

the Proceedings of the 2008 International Association of Maritime Economists

(IAME) Conference, 1-19, Dalian, China, April.

Carbone, Valentina, and Marcella De Martino. 2003. The Changing Role of Ports in

Supply-Chain Management: An Empirical Analysis. Maritime Policy & Management

30 (4): 305-320.

Carbone, Valentino, and Elisabeth Gouvernal. 2007. "Supply Chain and Supply Chain

Management: Appropriate Concepts for Maritime Studies." In Ports, Cities, and

Global Supply Chains, edited by Wang, James, Daniel Olivier, Theo E. Notteboom

and Brian Slack, 11-26. Burlington: Ashgate

Christopher, Martin. 2005. Logistics and Supply Chain Management: Creating Value-

Adding Networks. 3rd ed. New York: Financial Times Prentice Hall.

Constable, Gordon K., and D. Clay Whybark. 1978. The Interaction of Transportation

and Inventory Decisions. Decision Sciences 9 (4): 688-699.

Coyle, John J., Edward J. Bardi, and C. John Langley. 2009. Supply Chain Management:

A Logistics Perspective. Canada: South-Western Cengage Learning.

Drewry. 2010. Schedule Reliability Insight Q3 2010. London: Drewry Shipping

Consultants.

Drewry. 2012. Carrier Performance Insight April 2012. London: Drewry Shipping

Consultants.

Dullaert, Wout, Bert Vernimmen, El-Houssaine Aghezzaf, and Birger Raa. 2007.

Revisiting Service-Level Measurement for an Inventory System with Different

Transport Modes. Transport Reviews 27 (3): 273-283.

Fremont, A. 2007. Global Maritime Networks: The Case of Maersk. Journal of Transport

Geography 15 (6): 431-442.

Huang, George Q., X. Y. Zhang, and L. Liang. 2005. Towards Integrated Optimal

Configuration of Platform Products, Manufacturing Processes, and Supply Chains.

Journal of Operations Management 23: 267-290.

Lam, Jasmine Siu Lee, and Eddy Van De Voorde. 2011. Scenario Analysis for Supply

Chain Integration in Container Shipping. Maritime Policy & Management 38 (7):

705-725.

Lam, Jasmine Siu Lee, and Wei Yim Yap. 2011a. Container Port Competition and

Complementarity in Supply Chain Systems: Evidence from the Pearl River Delta.

Maritime Economics and Logistics 13 (2): 102-120.

18

Lam, Jasmine Siu Lee, and Wei Yim Yap. 2011b. Dynamics of Liner Shipping Network

and Port Connectivity in Supply Chain Systems: Analysis on East Asia. Journal of

Transport Geography 19 (6): 1272-1281.

Lam, Jasmine Siu Lee, Wei Yim Yap, and Kevin Cullinane. 2007. Structure, Conduct

and Performance on the Major Liner Shipping Routes. Maritime Policy &

Management 34 (4): 359-381.

Leach, Peter T. 2011. "Rival Carriers Ready Response to Daily Maersk." The Journal of

Commerce. Accessed 15 February 2012. http: //www.joc.com/rival-carriers-ready-

response-daily-maersk.

Leachman, Robert C. 2008. Port and Modal Allocation of Waterborne Containerized

Imports from Asia to the United States. Transportation Research Part E 44 (2): 313-

331.

Lee, Eon-Seong, and Dong-Wook Song. 2010. Knowledge Management for Maritime

Logistics Value: Discussing Conceptual Issues. Maritime Policy & Management 37

(6): 563-583.

Liberatore, Matthew J., and Tan Miller. 1995. A Decision Support Approach for

Transport Carrier and Mode Selection. Journal of Business Logistics 16 (2): 85-115.

Lovell, Antony, Richard Saw, and Jennifer Stimson. 2005. Product Value-Density:

Managing Diversity through Supply Chain Segmentation. International Journal of

Logistics Management 16 (1): 142-158.

Maersk Line. 2011. "Daily Maersk: Introducing Absolute Reliability." Accessed 15

February 2012. http://www.dailymaersk.com/.

Maersk Line. 2012. "Daily Maersk: Celebrating One Year of Reliability." Accessed 11

January 2013.

http://www.maerskline.com/link/?page=news&path=/news/story_page/12/D_Maersk.

Mangan, John, Chandra Lalwani, and Brian Fynes. 2008. Port-Centric Logistics.

International Journal of Logistics Management 19 (1): 29-41.

Marlow, Peter B., and Ana C. Paixao. 2003. Measuring Lean Ports Performance.

International Journal of Transport Management 1 (4): 189-202.

Midoro, Renato, and Alessandro Pitto. 2000. A Critical Evaluation of Strategic Alliances

in Liner Shipping. Maritime Policy & Management 27 (1): 31-40.

Moon, Daniel S. H. 2012. "What's Next?" Accessed 8 April 2012.

http://www.cee.ntu.edu.sg/Events/Pages/MaritimeEvents.aspx.

Notteboom, Theo E. 2006. The Time Factor in Liner Shipping Services. Maritime

Economics and Logistics 8 (1): 19-39.

Notteboom, Theo E., and Jean-Paul Rodrigue. 2008. Containerisation, Box Logistics and

Global Supply Chains: The Integration of Ports and Liner Shipping Networks.

Maritime Economics and Logistics 10 (1-2): 152-174.

Page, Paul. 2010. "Reality Shows." The Journal of Commerce. Accessed 10 January

2013. http://www.joc.com/reality-shows_20101210.html.

Paixao, Ana Cristina, and Peter Bernard Marlow. 2003. Fourth Generation Ports--a

Question of Agility? International Journal of Physical Distribution & Logistics

Management 33 (4): 355-376.

Panayides, P. M., and D. W. Song. 2008. Evaluating the Integration of Seaport Container

Terminals in Supply Chains. International Journal of Physical Distribution &

Logistics Management 38 (7): 562-584.

19

Panayides, Photis M., and Dong-Wook Song. 2009. Port Integration in Global Supply

Chains: Measures and Implications for Maritime Logistics. International Journal of

Logistics: Research & Applications 12 (2): 133-145.

Pawlik, T., L. Stemmler, A. J. Baird, and M. Helch. 2011. The Value of Container

Terminal Investment to Ocean Carrier Strategy. Maritime Economics & Logistics 13

(3): 319-341.

Rao, Ramadas. 2012. "Reliability Bugle Sounded." Accessed 10 January 2013.

http://www.maritimegateway.com/mgw/index.php?option=com_content&view=articl

e&id=530:rel.

Robinson, Ross. 2002. Ports as Elements in Value-Driven Chain Systems: The New

Paradigm. Maritime Policy & Management 29 (3): 241-255.

Robinson, Ross. 2006. Port-Oriented Landside Logistics in Australian Ports: A Strategic

Framework. Maritime Economics & Logistics 8 (1): 40-59.

Rodrigue, Jean-Paul, and Theo E. Notteboom. 2009. The Terminalization of Supply

Chains: Reassessing the Role of Terminals in Port/Hinterland Logistical

Relationships. Maritime Policy & Management 36 (2): 165-183.

Saldanha, John P., Dawn M. Russell, and John E. Tyworth. 2006. A Disaggregate

Analysis of Ccean Carriers' Transit Time Performance. Transportation Journal 45

(2): 39-60.

Saldanha, John P., John E. Tyworth, Peter F. Swan, and Dawn M. Russell. 2009. Cutting

Logistics Costs with Ocean Carrier Selection. Journal of Business Logistics 30 (2):

175-195.

Silver, Ea, Df Pyke, and R Peterson. 1998. Inventory Management and Production

Planning and Scheduling. 3rd ed. New York: John Wiley & Sons.

Slack, Brian, Claude Comtois, and Robert Mccalla. 2002. Strategic Alliances in the

Container Shipping Industry: A Global Perspective. Maritime Policy & Management

29 (1): 65-76.

Slack, Brian, and Antoine Fremont. 2005. Transformation of Port Terminal Operations:

From the Local to the Global. Transport Reviews 25 (1): 117-130.

Song, Dong-Wook, and Photis M. Panayides. 2008. Global Supply Chain and

Port/Terminal: Integration and Competitiveness. Maritime Policy & Management 35

(1): 73-87.

Szakonyi, Mark. 2011. "Shipping's New World, Grand Alliances to Merge Asia-Europe

Services." The Journal of Commerce. Accessed 15 February 2012.

http://www.joc.com/container-lines/new-world-grand-alliances-merge.

Tongzon, Jose, Young-Tae Chang, and Sang-Yoon Lee. 2009. How Supply Chain

Oriented Is the Port Sector? International Journal of Production Economics 122 (1):

21-34.

Train, Kenneth, and Wesley W. Wilson. 2008. Estimation on Stated-Preference

Experiments Constructed from Revealed-Preference Choices. Transportation

Research Part B 42 (3): 191-203.

Tyworth, John E., and Amy Zhaohui Zeng. 1998. Estimating the Effects of Carrier

Transit-Time Performance on Logistics Cost and Service. Transportation Research

Part A 32A (2): 89-97.

20

Vernimmen, Bert, Wout Dullaert, and Steve Engelen. 2007. Schedule Unreliability in

Liner Shipping: Origins and Consequences for the Hinterland Supply Chain. Maritime

Economics & Logistics 9 (3): 193-213.

Vidal, Carlos J., and Marc Goetschalckx. 2000. Modeling the Effect of Uncertainties on

Global Logistics Systems. Journal of Business Logistics 21 (1): 95-120.

Zamparini, Luca, John Layaa, and Wout Dullaert. 2011. Monetary Values of Freight

Transport Quality Attributes: A Sample of Tanzanian Firms. Journal of Transport

Geography 19 (6): 1222-1234.

Zhang, Abraham, and George Q. Huang. 2012. Impacts of Business Environment

Changes on Global Manufacturing Outsourcing in China. Supply Chain Management:

An International Journal 17 (2): 138-151.

Zhang, Abraham, Jasmine Siu Lee Lam, and George Q. Huang. 2013. Port Strategy in the

Era of Supply Chain Management: The Case of Hong Kong. Maritime Policy &

Management: (In press).

Appendix: Approximate transport lead time distributions

Transport lead times in days were used for deriving histograms.

Transport stage 1 (export drayage and port-of-origin dwell)

Service I: Histogram Min = 3, Max = 7, Freq = {1, 2, 4, 5, 3, 2, 1, 1, 1}

Service II: Histogram Min = 2.5, Max = 4.5, Freq = {4, 7, 5, 3, 1} Service III: Histogram

Min = 3, Max = 6, Freq = {1, 2, 5, 6, 4, 1, 1}

Daily Maersk: Histogram Min = 0.1, Max = 1.2, Freq = {2, 3, 5, 3, 2, 1, 1, 1, 1, 0, 0, 1}

Transport stage 3 (port-of-destination dwell and import drayage)

Services I ~ III: Histogram Min = 1, Max = 4, Freq = {1, 2, 4, 6, 4, 2, 1}

Daily Maersk: Histogram Min = 0.5, Max = 3, Freq = {3, 4, 6, 4, 2, 1}